arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computational Theory and Mathematics

Machine Learning-Based\nToxicological Modeling for\nScreening Environmental Obesogens

Siying Wu (7055027), Linping Wang (6397439), Daniel Schlenk (463101), Jing Liu (38537)

· Figshare

See on the map

Abstract

The emerging presence of environmental obesogens, chemicals\nthat\ndisrupt energy balance and contribute to adipogenesis and obesity,\nhas become a major public health challenge. Molecular initiating events\n(MIEs) describe biological outcomes resulting from chemical interactions\nwith biomolecules. Machine learning models based on MIEs can predict\ncomplex toxic end points due to chemical exposure and improve the\ninterpretability of models. In this study, a system was constructed\nthat integrated six MIEs associated with adipogenesis and obesity.\nThis system showed high accuracy in external validation, with an area\nunder the receiver operating characteristic curve of 0.78. Molecular\nhydrophobicity (SlogP_VSA) and direct electrostatic interactions (PEOE_VSA)\nwere identified as the two most critical molecular descriptors representing\nthe obesogenic potential of chemicals. This system was further used\nto predict the obesogenic effects of chemicals on the candidate list\nof substances of very high concern (SVHCs). Results from 3T3-L1 adipogenesis\nassays verified that the system correctly predicted obesogenic or\nnonobesogenic effects of 10 of the 12 SVHCs tested, and identified\nfour novel potential obesogens, including 2-benzotriazol-2-yl-4,6-di<i>tert</i>-butylphenol (UV-320), 4-(1,1,5-trimethylhexyl)phenol\n(p262-NP), 2-[4-(1,1,3,3-tetramethylbutyl)phenoxy]ethanol (OP1EO)\nand endosulfan. These validation data suggest that the screening system\nhas good performance in adipogenic prediction.

More like this